Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

20
12/07/2008 UAI 2008 Cumulative Distribution Networks and the Derivative- Sum-Product Algorithm Jim C. Huang and Brendan J. Frey Probabilistic and Statistical Inference Group, Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada

description

Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm. Jim C. Huang and Brendan J. Frey - PowerPoint PPT Presentation

Transcript of Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

Page 1: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Cumulative Distribution Networks and the Derivative-

Sum-Product Algorithm

Jim C. Huang and Brendan J. FreyProbabilistic and Statistical Inference Group, Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada

Page 2: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

• Problems where density models may be intractable

• e.g.: Modelling arbitrary dependencies

• e.g.: Modelling stochastic orderings

• Cumulative distribution network (CDN)

Motivation

e.g.: Predicting game outcomes in Halo 2

Page 3: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Cumulative distribution networks (CDNs)

• Graphical model of the cumulative distribution function (CDF)

• Example:

Page 4: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Positive convergence

Negative convergence

Monotonicity

Cumulative distribution functions

• Marginalization maximization

• Conditioning differentiation

Page 5: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Necessary/sufficient conditions on CDN

functions• Negative convergence (necessity and sufficiency):

• Positive convergence (sufficiency):

For each Xk, at least one neighboring function 0

All functions 1

Page 6: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Necessary/sufficient conditions on CDN functions

• Monotonicity lemma (sufficiency):

All functions monotonically non-decreasing…

Sufficient condition for a valid joint CDF: Each CDN function can

be a CDF of its arguments

Page 7: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Marginal independence• Marginalization maximization

– e.g.: X is marginally independent of Y

Page 8: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Conditional independence• Conditioning differentiation

– e.g.: X and Y are conditionally dependent given Z

– e.g.: X and Y are conditionally independent given Z

• Conditional independence No paths contain observed variables

Page 9: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Check:

A toy example

Markov random fields

Required “Bayes net”

Page 10: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Inference by message passing

• Conditioning differentiation

• Replace sum in sum-product with differentiation

• Recursively apply product rule via message-passing with messages ,

• Derivative-Sum-Product (DSP)

Page 11: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Derivative-sum-product

• In a CDN:

• In a factor graph:

Page 12: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Ranking in multiplayer gaming• e.g.: Halo 2 game with 7 players, 3 teams

Player skill functions

Player performance

Team performance

Given game outcomes, update player skills as a function of all player/team performances

Page 13: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Ranking in multiplayer gaming

= Local cumulative model linking team rank rn

with player performances xn

e.g.: Team 2 has rank 2

Page 14: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Ranking in multiplayer gaming

Enforce stochastic orderings between teams via h

= Pairwise model of team ranks rn,rn+1

Page 15: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

• CDN functions = Gaussian CDFs

• Skill updates:

• Prediction:

Ranking in multiplayer gaming

Page 16: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Results• Previous methods for ranking players:

– ELO (Elo, 1978)– TrueSkill (Graepel, Minka and Herbrich, 2006)

• After message-passing…

Page 17: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Summary• The CDN as a graphical model for CDFs

• Unique conditional independence structure

• Marginalization maximization

• Global normalization can be enforced locally

• Conditioning differentiation

• Efficient inference with Derivative-Sum-Product

• Application to Halo 2 Beta Dataset

Page 18: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Discussion• Need to be careful when applying to ordinal discrete variables…

• Principled method for learning CDNs

• Variational principle? (loopy DSP seems to work well)

• Future applications to – Hypothesis testing– Document retrieval– Collaborative filtering– Biological sequence search– …

Page 19: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Thanks

• Questions?

Page 20: Cumulative Distribution Networks and the Derivative-Sum-Product Algorithm

12/07/2008UAI 2008

Interpretation of skill updates

• For any given player let denote the outcomes of games he/she has played previously

• Then the skill function corresponds to